Mathias Gibbens <[email protected]> writes: > On Thu, 2026-08-13 at 20:16 +0200, Simon Josefsson wrote: >> What do you think, is it useful to do mass-pruning of packages like >> this? > > I'll add my "+1" to the others who have already replied. Thanks for > working on this!
Thanks for support! > In the past I think there's been friction trying to remove team- > maintained packages due to the effort it would take to re-introduce a > package, but with the updated NEW queue packaging, that's not really a > concern any more. Right -- my old preference has always been to leave things in unstable (at least) because getting packages into Debian was such a big hurdle. Today I think we can/should be more aggressive to remove packages, because in the (most likely) really rare situation, where a package end up being needed later on, is not problematic any more. Having stuff in unstable/testing that nobody uses or cares about has a cost. For some packages I've looked that had recent QA uploads I would have prefered if that QA time were spent on packages that actually have some use inside Debian. I've made some of these QA uploads myself (for the go.mod addition), and realized it was pointless to spend time updating a golang-*-dev package that doesn't have any consumers inside Debian. > My one suggestion would be that no automated script should be filing > RM bugs within ~3 months of the expected soft freeze of a Debian > release. That should hopefully prevent any last-minute surprise > removals that may not allow sufficient time for a package to be re- > introduced for an upcoming release. Good point. Let's do this now in the next couple of months. /Simon
signature.asc
Description: PGP signature
